A processing method for cloud service platform

Abstract

The present application relates to the field of electronic digital data processing technology, and in particular to a processing method for a cloud service platform. The method comprises: obtaining user attribute information A of the cloud service platform; traversing A, and according to p i ,q i and i Get the time period e that the ith user of the cloud service platform intends to process data i According to p i and e i All users of cloud service platform services are classified to classify users into z categories; z is a preset initial classification number; the minimum value of the difference in processing time between categories corresponding to classifying all users of cloud service platform services into z categories is obtained; if classifying all users of cloud service platform services into z categories can meet the preset classification conditions, computing resources of the cloud service platform are allocated to users of cloud service platform services according to the classification result of classifying all users of cloud service platform services into z categories. The present invention improves the overall processing speed of the cloud service platform in processing user requests.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a processing method of a cloud service platform. Background Art

[0002] The cloud service platform is a platform built on cloud computing technology. It can provide computing services to users and allow users to access shared computing resources on the Internet. The cloud service platform usually allocates computing resources according to user needs. When a user sends a computing resource request to the cloud computing platform, the cloud computing service platform allocates computing resources to the user according to the user's request. However, there are performance differences in the computing resources of the cloud service platform. The computing resources with better performance have faster processing speed. How to reasonably allocate the computing resources of the cloud service platform and thus improve the overall processing speed of the cloud service platform in processing user requests is an urgent problem to be solved. Summary of the invention

[0003] The present invention aims to provide a processing method for a cloud service platform to reasonably allocate computing resources of the cloud service platform and improve the overall processing speed of the cloud service platform in processing user requests.

[0004] According to the present invention, a method for processing a cloud service platform is provided, the method comprising the following steps:

[0005] S100, obtaining user attribute information A of the cloud service platform, A=(a1, a2, ..., a i ,…,a n ), a i The attribute information of the i-th user served by the cloud service platform. The value of i ranges from 1 to n, and n is the number of users served by the cloud service platform; a i =(p i ,q i ,s i ), p i The type of data processing to be performed by the ith user of the cloud service platform, q i The amount of data to be processed by the i-th user of the cloud service platform, s i The start time of data processing for the ith user of the cloud service platform.

[0006] S200, traverse A, according to a i Included p i ,q i and i Get the time period e that the ith user of the cloud service platform intends to process data i .

[0007] S300, according to p i and ei All users of the cloud service platform are classified into z categories of users; z is a preset initial number of categories.

[0008] S400, obtaining a minimum value of the inter-class processing time difference corresponding to when all users of the cloud service platform are divided into z classes of users.

[0009] S500, if classifying all users of the cloud service platform service into z categories of users can meet the preset classification conditions, then the computing resources of the cloud service platform are allocated to the users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z categories of users; the preset classification conditions include that the minimum value of the difference in processing time between classes is greater than or equal to a preset difference threshold.

[0010] Compared with the prior art, the present invention has at least the following beneficial effects:

[0011] The present invention obtains the attribute information of all users of the cloud service platform, and the attribute information of each user includes the data processing type to be performed by the user, the amount of data to be processed and the start time of the data processing to be performed; based on the attribute information of each user, the present invention obtains the time period for each user to perform data processing, and classifies the users of the cloud service platform service in combination with the data processing type to be performed by each user, so that the data processing type and the time period for data processing to be performed by users in the same class are relatively similar; if the sub-classification can meet the preset classification conditions, the computing resources are allocated to the users according to the sub-classification results; the preset classification conditions of the present invention include that the minimum value of the processing time difference between classes is greater than or equal to the preset difference threshold, which can make the processing time period corresponding to each class differ greatly, which is conducive to the cloud service platform to allocate computing resources with better performance to different classes in different time periods; moreover, in this embodiment, the data processing type to be performed by each class of users is relatively similar, and the computing resources with better performance of the cloud service platform are used to concentrate on the processing of the calculations with relatively consistent types, which is conducive to further improving the processing speed of the computing resources with better performance to process user requests; thus, the present invention realizes the reasonable allocation of computing resources of the cloud service platform and improves the overall processing speed of the cloud service platform to process user requests. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1A flowchart of a processing method for a cloud service platform provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0015] According to this embodiment, Figure 1 As shown, a processing method for a cloud service platform is provided, the method comprising the following steps:

[0016] S100, obtaining user attribute information A of the cloud service platform, A=(a1, a2, ..., a i ,…,a n ), a i The attribute information of the i-th user served by the cloud service platform. The value of i ranges from 1 to n, and n is the number of users served by the cloud service platform; a i =(p i ,q i ,s i ), p i The type of data processing to be performed by the ith user of the cloud service platform, q i The amount of data to be processed by the i-th user of the cloud service platform, s i The start time of data processing for the ith user of the cloud service platform.

[0017] In this embodiment, the data processing types that different users of the cloud service platform intend to perform are the same or different. The data processing type that the user intends to perform refers to the type of data that the user intends to process using the computing resources of the cloud service platform. For example, some data types are attendance data, some data types are warehouse inventory management data, and some data types are vehicle dispatch data. The cloud service platform can pre-set multiple data types. The data processing processes corresponding to the same data type are relatively similar, and the data processing processes corresponding to different data types are quite different. Users can select the data type corresponding to the data they intend to process.

[0018] In this embodiment, the type of data processing to be performed by the user, the amount of data to be processed, and the start time of the data processing to be performed are information given by the user and are known information.

[0019] S200, traverse A, according to a i Included p i ,q i andi Get the time period e that the ith user of the cloud service platform intends to process data i .

[0020] In this embodiment, according to a i Included p i ,q i and i The time period e that the ith user of the cloud service platform intends to process data can be obtained i As a specific implementation, S200 includes:

[0021] S210, according to q i and p i Get the processing time L of the ith user of the cloud service platform i .

[0022] In this embodiment, the data processing type and processing amount that each user intends to perform are known, and the processing time of each user is obtained based on a pre-established relationship between the data processing type, data amount and processing time. Optionally, the relationship between the data processing type, data amount and processing time is pre-established based on experience.

[0023] In the pre-established relationship between data processing type, data volume and processing time, when the data processing type is the same, the processing time is positively correlated with the data volume; when the data volume is the same, the processing time is positively correlated with the processing complexity corresponding to the data processing type.

[0024] S220, get the same as s i The distance is L i and later than s i Time b i .

[0025] S230, will [s i ,b i ] Determine the time period for data processing for the i-th user served by the cloud service platform.

[0026] S300, according to p i and e i All users of the cloud service platform are classified into z categories of users; z is a preset initial number of categories.

[0027] As a specific implementation, z is an empirical value.

[0028] As a specific implementation, a k-means clustering method is used to divide all users of the cloud service platform into z categories of users.

[0029] As a specific implementation, S300 further includes:

[0030] S310: randomly select z users from n users served by the cloud service platform as z initial users.

[0031] S320, obtaining the distance d between the jth user and the δth initial user among the n users served by the cloud service platform except the zth initial users j,δ , d j,δ =w1×β j,δ +w2×(1-η j,δ / ε j,δ ), w1 and w2 are the weights of data processing type and data processing time period respectively, <w1<1,0<w2<1,w1+w2=1,β j,δ The difference between the data processing types to be performed by the jth user and the δth initial user among the n users serving the cloud service platform, 0≤β j,δ ≤1,η j,δ The overlapping duration of the time period during which the jth user, excluding the zth initial users, and the δth initial user intend to process data among the n users serving the cloud service platform, ε j,δ The total span of the time period during which the jth user and the δth initial user plan to process data among the n users served by the cloud service platform, excluding the z initial users; the value range of j is 1 to nz, and the value range of δ is 1 to z.

[0032] In this embodiment, a difference value between any two data processing types is pre-established. Optionally, the difference value between any two data processing types is an empirical value.

[0033] In this embodiment, if the time periods for data processing that two users intend to perform do not overlap, the overlapping duration of the time periods for data processing corresponding to the two users is 0, and the total span duration of the time periods for data processing corresponding to the two users is the sum of the durations of the time periods for data processing corresponding to the two users; if the time periods for data processing that two users intend to perform overlap, the total span duration of the time periods for data processing corresponding to the two users is the duration between the earlier start time of the start time of the data processing that the two users intend to perform and the later end time of the end time of the data processing that the two users intend to perform.

[0034] S330, classify the other users except the z initial users among the n users served by the cloud service platform into the class corresponding to the initial users closest to them.

[0035] In this embodiment, after S330, it also includes: updating the cluster center; judging whether the preset stop condition is met, and if so, outputting the cluster result; otherwise, performing the next round of iteration. In this embodiment, the process of determining the cluster center of a class includes: determining the data processing type with the most occurrences among the data processing types of all users included in the class as the data processing type of the cluster center of the class, determining the mean of the start time of the data processing time period of the cluster center of the class for all users included in the class as the start time of the data processing time period of the cluster center of the class, and determining the mean of the end time of the data processing of all users included in the class as the end time of the data processing time period of the cluster center of the class; the cluster center of the class can be determined based on the data processing type of the cluster center of the class, the start time and the end time of the data processing time period.

[0036] As a specific implementation method, the method for determining the distance between two users and the method for determining the distance between a certain user and the cluster center in the clustering process are similar to the method for obtaining d j,δ The method is similar and will not be described here.

[0037] As a specific implementation, the preset stop condition is that the cluster center is no longer updated or the number of iterations reaches a preset number.

[0038] S400, obtaining a minimum value of the inter-class processing time difference corresponding to when all users of the cloud service platform are divided into z classes of users.

[0039] As a specific implementation, S400 includes:

[0040] S410, obtaining a classification result G of classifying all users of the cloud service platform into z types of users, G=(g1, g2, ..., g x ,…,g z ), g x The xth class is obtained by dividing all users of the cloud service platform, and the value of x ranges from 1 to z.

[0041] S420, get g x The mean start time m - x , m - x =avg(h - x,1 ,h - x,2 ,…,h - x,r(x) ,…,h - x,R(x) ), h - x,r(x) g xThe starting time of the data processing of the r(x)th user is included, avg( ) is the average value; the value range of r(x) is 1 to R(x), R(x) is g x The number of users included.

[0042] S430, get g x The mean end time m + x , m + x =avg(h + x,1 ,h + x,2 ,…,h + x,r(x) ,…,h + x,R(x) ), h + x,r(x) g x The end time of data processing to be performed by the included r(x)th user.

[0043] As a specific implementation, h + x,r(x) According to h - x,r(x) and g x The processing time of the r(x)th user is obtained.

[0044] S440, sort G in descending order according to the corresponding start time averages to obtain G', G'=(g'1, g'2, ..., g' x ,…,g' z ), g' x is the class with the xth earliest mean start time in G.

[0045] In this embodiment, the class with the 1st earliest corresponding mean start time in G is the class with the earliest corresponding mean start time in G, the class with the 2nd earliest corresponding mean start time in G is the class with the second earliest corresponding mean start time in G, and so on. The class with the z-1th earliest corresponding mean start time in G is the class with the second latest corresponding mean start time in G, and the class with the zth earliest corresponding mean start time in G is the class with the latest corresponding mean start time in G.

[0046] S450, (Δt2,…,Δt y ,…,Δt z ) is determined as the minimum value of the inter-class processing time difference corresponding to the division of all users of the cloud service platform into z classes of users; Δt y for g' y The mean start time of g' y-1The time interval between the mean end times of , the value of y ranges from 2 to z.

[0047] S500, if classifying all users of the cloud service platform service into z categories of users can meet the preset classification conditions, then the computing resources of the cloud service platform are allocated to the users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z categories of users; the preset classification conditions include that the minimum value of the difference in processing time between classes is greater than or equal to a preset difference threshold.

[0048] As a specific implementation, the method of allocating computing resources of the cloud service platform to users of the cloud service platform according to the classification result of classifying all users of the cloud service platform into z types of users includes:

[0049] S510, obtaining a classification result G of classifying all users of the cloud service platform into z types of users, G=(g1, g2, ..., g x ,…,g z ), g x The xth class is obtained by dividing all users of the cloud service platform, and the value of x ranges from 1 to z.

[0050] S520, according to g x The time period in which the user intends to process the data is obtained x Corresponding processing time period T x .

[0051] As a specific implementation, S520 includes:

[0052] S521, get g x The time period H during which the user intends to process the data x , H x =(h x,1 ,h x,2 ,…,h x,r(x) ,…,h x,R(x) ), h x,r(x) g x The time period during which the r(x)th user intends to process data, r(x) ranges from 1 to R(x), and R(x) is g x The number of users included; h x,r(x) =(h - x,r(x) ,h + x,r(x) ), h - x,r(x) g x The start time of data processing for the r(x)th user, h + x,r(x) g xThe end time of data processing to be performed by the included r(x)th user.

[0053] S522, get g x The corresponding earliest start time gt - x With g x The corresponding latest end time gt + x gt - x =min(h - x,1 ,h - x,2 ,…,h - x,r(x) ,…,h - x,R(x) ),gt + x =max(h + x,1 ,h + x,2 ,…,h + x,r(x) ,…,h + x,R(x) ), min() is the earliest time, and max() is the latest time.

[0054] S523, if gt - x No earlier than g x The latest end time of the previous category and gt + x No later than g x The earliest end time of the latter category, then gt - x Determined as T x The start time of gt + x Determined as T x End time of x The previous category is the average start time ratio g in the z category obtained by division x The average start time is earlier and similar to g x The class with the closest average start time; g x The latter category is the average start time ratio g in the z category obtained by division x The average start time is later than that of g x The class with the closest average start time.

[0055] In this embodiment, g' x The first category is g' x-1 , g' x The latter type is g'x+1 If g' x If g'1 is - x Determined as T x The start time; if g' x for g' z , then gt + x Determined as T x The end time of .

[0056] In this embodiment, S523 further includes: if gt - x Earlier than g x The latest end time of the previous category, then enter S524; if gt + x Later than g x If the earliest start time of the latter category is selected, the process proceeds to S526.

[0057] S524, get g x The latest end time of the previous class.

[0058] S525, g x The latest end time of the previous category is the same as gt - x The mean value of is determined as T x The start time of the

[0059] S526, get g x The earliest start time of the latter category.

[0060] S527, g x The earliest start time of the latter category is the same as gt + x The mean value of is determined as T x The end time of .

[0061] In this embodiment, if gt - x Earlier than g x The latest end time of the previous category and gt + x Later than g x The earliest start time of the latter category, then g x The latest end time of the previous category is the same as gt - x The mean value of is determined as T x The start time and g x The earliest start time of the latter category is the same as gt + x The mean value of is determined as T x The end time of .

[0062] S530, in T x Inside, for g x The method includes allocating a first computing resource of a cloud service platform to a user, wherein the priority of the first computing resource is higher than the priority of a second computing resource of the cloud service platform.

[0063] In this example, the priority of the first computing resource of the cloud service platform is higher than the priority of the second computing resource of the cloud service platform. x In, priority is g x The user allocates the first computing resource of the cloud service platform, if the first computing resource of the cloud service platform is less than g x The first computing resources required by the users included in the cloud service platform will be allocated according to the start time of the data processing to give priority to the first computing resources for users with earlier start time of data processing; if the first computing resources of the cloud service platform are greater than g x The computing resources required by the users included in T x If the data is also processed in T, the redundant first computing resources are allocated to other classes. x The user who starts processing data within .

[0064] As a specific implementation, the preset classification condition only includes that the minimum value of the difference in processing time between classes is greater than or equal to the preset difference threshold; as another specific implementation, the preset classification condition also includes that the variance of the difference in processing time between classes is less than or equal to the preset variance threshold, thereby, the time interval for the first computing resource of the cloud service platform to transfer from processing one type of user to another type of user is more balanced. In this embodiment, the variance of the difference in processing time between classes corresponding to G is (Δt2,…,Δt y ,…,Δt z ) includes the variance of the time interval.

[0065] As a specific implementation, S500 further includes: if classifying all users of the cloud service platform service into category z users cannot meet the preset classification condition, then entering S600.

[0066] S600: Classify all users of the cloud service platform into z+1 categories of users.

[0067] S700: If classifying all users of the cloud service platform service into z+1 category users can meet the preset classification conditions, then the computing resources of the cloud service platform are allocated to the users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z+1 category users; otherwise, enter S800.

[0068] In this embodiment, the method of classifying all users of the cloud service platform service into z+1 category users is similar to the above-mentioned method of classifying all users of the cloud service platform service into z category users, and will not be repeated here.

[0069] In this embodiment, when allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z+1 categories of users, the allocation method is similar to the method of allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z categories of users. It also allocates the first computing resources of the cloud service platform to users in the class within the corresponding processing time period of each class, which will not be repeated here.

[0070] S800, if classifying all users of the cloud service platform service into z-1 category users can meet the preset classification conditions, then according to the classification result of classifying all users of the cloud service platform service into z-1 category users, the computing resources of the cloud service platform are allocated to the users of the cloud service platform service.

[0071] In this embodiment, the method of classifying all users of the cloud service platform service into z-1 category users is similar to the above-mentioned method of classifying all users of the cloud service platform service into z category users, and will not be repeated here.

[0072] In this embodiment, when allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z-1 category users, the allocation method is similar to the method of allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z category users. It also allocates the first computing resources of the cloud service platform to users in the class within the corresponding processing time period of each category, which will not be repeated here.

[0073] As a specific implementation, S800 further includes: if classifying all users of the cloud service platform service into category z-1 users cannot meet the preset classification condition, then entering S900.

[0074] S900, divide all users of the cloud service platform into z+2 categories of users.

[0075] S910, if classifying all users of the cloud service platform service into z+2 category users can meet the preset classification conditions, then allocate computing resources of the cloud service platform to the users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z+2 category users; otherwise, enter S920.

[0076] In this embodiment, the method of classifying all users of the cloud service platform service into z+2 types of users is similar to the above-mentioned method of classifying all users of the cloud service platform service into z types of users, which will not be repeated here.

[0077] In this embodiment, when allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z+2 categories of users, the allocation method is similar to the method of allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z categories of users. It also allocates the first computing resources of the cloud service platform to users in the class within the corresponding processing time period of each class, which will not be repeated here.

[0078] S920: If classifying all users of the cloud service platform service as class z-2 users can meet the preset classification conditions, computing resources of the cloud service platform are allocated to the users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service as class z-2 users.

[0079] In this embodiment, the method of classifying all users of the cloud service platform service into z-2 category users is similar to the above-mentioned method of classifying all users of the cloud service platform service into z category users, and will not be repeated here.

[0080] In this embodiment, when allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z-2 categories of users, the allocation method is similar to the method of allocating computing resources of the cloud service platform to users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z categories of users. It also allocates the first computing resources of the cloud service platform to users in the class within the corresponding processing time period of each class, which will not be repeated here.

[0081] In this embodiment, if dividing all users of the cloud service platform into z+2 and z-2 categories fails to meet the preset classification conditions, then all users of the cloud service platform will be divided into z+3 or z-3 categories, and so on, until the division meets the preset classification conditions or the preset number of divisions is reached; if the preset number of divisions is reached, the division result corresponding to the maximum inter-class processing time difference will be used as the basis for allocating the first computing resources of the cloud service platform.

[0082] This embodiment obtains the attribute information of all users of the cloud service platform service, and the attribute information of each user includes the data processing type to be performed by the user, the amount of data to be processed, and the start time of the data processing to be performed; based on the attribute information of each user, this embodiment obtains the time period for each user to perform data processing, and classifies the users of the cloud service platform service in combination with the data processing type to be performed by each user, and the data processing type and the time period for the data processing to be performed by the users in the same class are relatively similar; if the sub-classification can meet the preset classification conditions, the computing resources are allocated to the users according to the sub-classification results; the preset classification conditions of this embodiment include that the minimum value of the processing time difference between classes is greater than or equal to the preset difference threshold, which can make the processing time period corresponding to each class differ greatly, which is conducive to the cloud service platform to allocate computing resources with better performance to different classes in different time periods; moreover, in this embodiment, the data processing type to be performed by each class of users is relatively similar, and the computing resources with better performance of the cloud service platform are used to centrally process relatively consistent calculations, which is conducive to further improving the processing speed of the computing resources with better performance to process user requests; thus, this embodiment realizes the reasonable allocation of computing resources of the cloud service platform and improves the overall processing speed of the cloud service platform to process user requests.

[0083] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A processing method for a cloud service platform, characterized in that: The method comprises the following steps: S100, obtaining user attribute information A of the cloud service platform, A=(a1, a2, ..., a i ,…,a n ), a i The attribute information of the i-th user served by the cloud service platform. The value of i ranges from 1 to n, and n is the number of users served by the cloud service platform; a i =(p i ,q i ,s i ), p i The type of data processing to be performed by the ith user of the cloud service platform, q i The amount of data to be processed by the i-th user of the cloud service platform, s i The start time of data processing for the ith user of the cloud service platform; S200, traverse A, according to a i Included p i ,q i and i Get the time period e that the i-th user of the cloud service platform intends to process data i ; S300, according to p i and e i Classify all users of the cloud service platform service to divide all users of the cloud service platform service into z categories of users; z is a preset initial classification number; S400, obtaining the minimum value of the difference in processing time between classes corresponding to when all users of the cloud service platform are divided into z classes of users; S500, if classifying all users of the cloud service platform service into z-type users can meet a preset classification condition, then allocating computing resources of the cloud service platform to the users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service into z-type users; the preset classification condition includes that the minimum value of the difference in processing time between classes is greater than or equal to a preset difference threshold; The S300 also includes: S310, randomly selecting z users from n users served by the cloud service platform as z initial users; S320, obtaining the distance d between the jth user and the δth initial user among the n users served by the cloud service platform except the zth initial users j,δ , d j,δ =w1×β j,δ +w2×(1-η j,δ / ε j,δ ), w1 and w2 are the weights of data processing type and data processing time period respectively, <w1<1,0<w2<1,w1+w2=1,β j,δ The difference between the data processing types to be performed by the jth user and the δth initial user among the n users serving the cloud service platform, 0≤β j,δ ≤1,η j,δ The overlapping duration of the time period during which the jth user, excluding the zth initial users, and the δth initial user intend to process data among the n users serving the cloud service platform, ε j,δ The total span of the time period during which the jth user and the δth initial user intend to process data among the n users served by the cloud service platform, excluding the zth initial users; the value range of j is 1 to nz, and the value range of δ is 1 to z; S330, classifying the other users except the z initial users among the n users served by the cloud service platform into the class corresponding to the initial users closest to them; S400 includes: S410, obtaining a classification result G of classifying all users of the cloud service platform into z types of users, G=(g1, g2, ..., g x ,…,g z ), g x The xth class is obtained by dividing all users of the cloud service platform, and the value of x ranges from 1 to z; S420, get g x The mean start time m - x , m - x =avg(h - x,1 ,h - x,2 ,…,h - x,r(x) ,…,h - x,R(x) ), h - x,r(x) g x The starting time of the data processing of the r(x)th user is included, avg( ) is the average value; the value range of r(x) is 1 to R(x), R(x) is g x The number of users included; S430, get g x The mean end time m + x , m + x =avg(h + x,1 ,h + x,2 ,…,h + x,r(x) ,…,h + x,R(x) ), h + x,r(x) g x The end time of data processing to be performed by the included r(x)th user; S440, sort G in descending order according to the corresponding start time averages to obtain G', G'=(g'1, g'2, ..., g' x ,…,g' z ), g' x is the class with the xth earliest mean start time in G; S450, (Δt2,…,Δt y ,…,Δt z ) is determined as the minimum value of the inter-class processing time difference corresponding to the division of all users of the cloud service platform into z classes of users; Δt y for g' y The mean start time of g' y-1 The time interval between the mean end times of , the value of y ranges from 2 to z.

2. The processing method of the cloud service platform according to claim 1, characterized in that: The method of allocating computing resources of the cloud service platform to users of the cloud service platform according to the classification result of classifying all users of the cloud service platform into z types of users includes: S510, obtaining a classification result G of classifying all users of the cloud service platform into z types of users, G=(g1, g2, ..., g x ,…,g z ), g x The xth class is obtained by dividing all users of the cloud service platform, and the value of x ranges from 1 to z; S520, according to g x The time period in which the user intends to process the data is obtained x Corresponding processing time period T x ; S530, in T x Inside, for g x The method includes allocating a first computing resource of a cloud service platform to a user, wherein the priority of the first computing resource is higher than the priority of a second computing resource of the cloud service platform.

3. The processing method of the cloud service platform according to claim 2, characterized in that: S520 includes: S521, get g x The time period H during which the user intends to process the data x , H x =(h x,1 ,h x,2 ,…,h x,r(x) ,…,h x,R(x) ), h x,r(x) g x The time period during which the r(x)th user intends to process data, r(x) ranges from 1 to R(x), and R(x) is g x The number of users included; h x,r(x) =(h - x,r(x) ,h + x,r(x) ), h - x,r(x) g x The start time of data processing for the r(x)th user, h + x,r(x) g x The end time of data processing to be performed by the included r(x)th user; S522, get g x The corresponding earliest start time gt - x With g x The corresponding latest end time gt + x gt - x =min(h - x,1 ,h - x,2 ,…,h - x,r(x) ,…,h - x,R(x) ), gt + x =max(h + x,1 ,h + x,2 ,…,h + x,r(x) ,…,h + x,R(x) ), min() is the earliest time, max() is the latest time; S523, if gt - x No earlier than g x The latest end time of the previous category and gt + x No later than g x The earliest end time of the latter category, then gt - x Determined as T x The start time of gt + x Determined as T x End time of x The previous category is the average start time ratio g in the z category obtained by division x The average start time is earlier and similar to g x The class with the closest average start time; g x The latter category is the average start time ratio g in the z category obtained by division x The average start time is later than that of g x The class with the closest average start time.

4. The processing method of the cloud service platform according to claim 3, characterized in that: S523 also includes: If gt - x Earlier than g x The latest end time of the previous category, then enter S524; if gt + x Later than g x The earliest start time of the latter category, then proceed to S526; S524, get g x The latest end time of the previous category; S525, g x The latest end time of the previous category is the same as gt - x The mean value of is determined as T x The start time of S526, get g x The earliest start time of the latter category; S527, g x The earliest start time of the latter category is the same as gt + x The mean value of is determined as T x The end time of .

5. The processing method of the cloud service platform according to claim 1, characterized in that: S500 also includes: if classifying all users of the cloud service platform service into class z users cannot meet the preset classification condition, entering S600; S600, classify all users served by the cloud service platform into z+1 categories of users; S700, if classifying all users of the cloud service platform service into z+1 category users can meet the preset classification condition, then according to the classification result of classifying all users of the cloud service platform service into z+1 category users, the computing resources of the cloud service platform are allocated to the users of the cloud service platform service; otherwise, enter S800; S800, if classifying all users of the cloud service platform service into z-1 category users can meet the preset classification conditions, then according to the classification result of classifying all users of the cloud service platform service into z-1 category users, the computing resources of the cloud service platform are allocated to the users of the cloud service platform service.

6. The processing method of the cloud service platform according to claim 1, characterized in that: S200 includes: S210, according to q i and p i Get the processing time L of the ith user of the cloud service platform i ; S220, get the same as s i The distance is L i and later than s i Time b i ; S230, will [s i ,b i ] Determine the time period for data processing for the i-th user served by the cloud service platform.

7. The processing method of the cloud service platform according to claim 5, characterized in that: S800 also includes: if classifying all users of the cloud service platform service into class z-1 users cannot meet the preset classification condition, entering S900; S900, classify all users of the cloud service platform into z+2 categories of users; S910, if classifying all users of the cloud service platform service into z+2 users can meet the preset classification condition, then according to the classification result of classifying all users of the cloud service platform service into z+2 users, the computing resources of the cloud service platform are allocated to the users of the cloud service platform service; otherwise, enter S920; S920: If classifying all users of the cloud service platform service as class z-2 users can meet the preset classification conditions, computing resources of the cloud service platform are allocated to the users of the cloud service platform service according to the classification result of classifying all users of the cloud service platform service as class z-2 users.

Images